Analysis of the quantitative assessment data was first conducted by CEEE. With each spatial skill as an outcome, several mixed models were conducted with the data, and Type III sums of squares were calculated to find omnibus effects.
Fixed effects included time, group (treatment vs. control), grade (centered at the middle grade if applicable), gender, and a time by group interaction as fixed factors. A random classroom effect was also included.
Additional quantitative analyses were conducted by Dr. Eric McLaughlin at the University of Redlands. Dr. McLaughlin replicated the analyses reported by CEEE and conducted additional analyses requested by the project leaders. These analyses probed the influences of sociodemographic variables, gender, grade level, extracurricular spatial activities, and other relevant variables.
The qualitative data analysis conducted by CEEE began with transcription of the interview audio recordings.
Next, three independent researchers read the interviews and noted key themes, or emergent codes, that came up repeatedly or on which there was consensus among the teachers/project leaders. Thus, the themes were developed inductively by these researchers based on the interview material.